REMINISCENCE - UNIVERSITY - PART 2

by BERNARD A HODSON

My previous article outlined the acquisition of a powerful computer at the University of Manitoba as a prelude to establishing a Department of Computer Science (CS). At the time CS was located in the Faculty of Graduate Studies, whose Dean was Dr. Lionel B Funt.

My initial strategy, apart from offering a variety of service courses to groups such as engineers and mathematicians, was to establish a Master's degree in CS, based on a thesis, along with a strong research thrust addressing some of the problems of the day. To do this I hired in both academic and support staff people with a wide variety of disciplines, people with vision. We then initiated a wide ranging research program which attracted significant funding, as well as curious researchers from abroad. The research provided topics for thesis work.

In what follows the reader should keep in mind that the computer at the time had initially 128k words of memory, later 256k words. Throughout IT history there have been cycles of small memory having to be used effectively, and I continue to be appalled at today's iniquitous waste of resources. In my own current research today I have developed what I call A New Kind of Computing where an easy to use numeric code structure defines applications and the actual conventional machine code is from 1k to 8k words, depending on the features used.

Dr. Carol Abraham joined us from CSIRO in Australia. He developed one of the early 'roll in, roll out' systems which enabled multiple applications to be run simultaneously (time sharing). Research was done on computer networks, with terminals or computers of various kinds attached in Montreal, the Atomic Energy Research facility in Pinawa, Manitoba, nursing stations at a local hospital, non IBM computers in the Medical School, on occasion at conferences in Vancouver, Banff and Houston, as well as an attachment to a Digital Equipment PDP9 which, in its turn, was controlling a Cyclotron in the Physics Department. This latter linkage was one of the early developments in 'real time'. The Cyclotron fed the PDP9 which filtered the data and passed it to the 360/65 for number crunching. Results were sent back to the PDP9 for visual display, so that physicists could modify the Cyclotron experiment in real time. All this was done while students were pushing through hundreds of programs, and all with the minimal memory mentioned earlier.

Dr. Brodin joined us from the French Aeronautical Institute, then located in Paris, now in Toulouse. Dr Jan Blatny came from being head of Radio Engineering at the Technical University of Brno in Czechoslovakia and returned to form a Department of CS, also building an IBM 360 compatible computer which was still functioning when I visited there 20 or so years later. Pamela Morgan did some work in computer assisted instruction helping two pathologists in the Medical School develop a computer assist for teaching haematology. Don Costin experimented with sppech recognition.

A lawyer, Steve Skelly, developed a system for computerising Manitoba Statutes while they were going through the legislature, matching them for compatibility with existing statutes. He also worked with a Florida based company on early desk top publishing (as it used to be called) and was later hired by the Federal department of Justice, rising to become senior Assistant Deputy Minister there. Richard Morgan, a classics scholar, used the computer try and decipher Minoan Seals, left intriguingly by an early Mediterranean civilisation. A staff member with a degree in German did early work on thesauruses. My own work included pioneering in hospital information systems (subject of later articles) and in generalised program development using extremely small memory. Karl Schmidt developed two dimensional presentations of three related variables and later worked with members of the Mathematics Department writing joint papers.

One item I definitely advocated was the ability of graduates to comprehend 'systems analysis', where multiple requirements have to be included in a suite of related application programs. To do this I developed a 'systems language'; which enabled students to create a 'system'. It was successfully used with students and then with senior managers at the Banff School of Advanced Management, as well as with senior managers in the USA. Macmillan of Canada asked if I could develop the notes given at Banff in to a book, which I did, the book also being published in the USA.

Teaching was not limited just to the University. I was asked by the CTV network to develop a series of 10 presentations for their University of the Air series, which I did (one of the ten was also used by United Airlines, in the USA, as part of their training program). CBC also asked me to develop four presentations for their Schools Broadcast network.